von Neumann entropy

# tag · @ author · ! scientist

Von Neumann entropy is the quantum analogue of classical Shannon entropy, but for systems governed by quantum rules. A quantum state is described by a density matrix—a table of probabilities. If the particle is in a pure state, the density matrix is simple, and the entropy is zero. Upon interaction with the environment, the state becomes mixed, the density matrix becomes more complex, and the entropy increases. The maximum value for a system of n qubits is n—complete uncertainty. This quantity underlies modern protocols for quantum data transmission and error correction.

History

In 1927, mathematician John von Neumann first described how to compute the degree of ignorance in the quantum world using a special number—entropy.

How it works

Imagine a deck of playing cards: if it is arranged by suits, entropy is low—it's easy to guess the next card. If the deck is well shuffled, entropy is high. Here, instead of cards, we have quantum states (like 'heads' and 'tails' for a particle), and instead of the order in the deck, the degree of quantum entanglement. Von Neumann entropy calculates how much information we lose when mixing quantum possibilities.

💡 Von Neumann entropy cannot decrease without exchanging information with the environment—just as a shuffled deck cannot be instantly ordered without knowing the exact arrangement of cards.

Von Neumann entropy, introduced by John von Neumann in 1927 and published in 'Mathematical Foundations of Quantum Mechanics', quantifies the lack of classical information about a quantum system. For a density matrix ρ, it is computed as S(ρ) = −Tr(ρ ln ρ), where Tr is the trace and ln is the matrix logarithm. In the eigenbasis, where ρ = ∑ λ_i |i⟩⟨i|, the entropy takes the familiar form S = −∑ λ_i ln λ_i with eigenvalues λ_i (probabilities). It is non-negative, zero for pure states, and reaches its maximum ln d for a completely mixed state of dimension d. In quantum information, it serves as a measure of entanglement: the entropy of a subsystem gives the entanglement entropy. Properties of additivity and strong subadditivity make it a central concept in the theory of quantum correlations.

History

Von Neumann introduced quantum entropy in 1932, drawing on analogies with statistical thermodynamics. The idea deepened after the birth of Shannon's information theory in 1948, when the connection with channel capacity became clear. In the 1970s–80s, in works on quantum thermodynamics and decoherence (Zurek, Lindblad), entropy became key to describing the transition from quantum to classical. David Deutsch in 1985 used it in his pioneering concepts of quantum computing, laying the foundations of quantum information theory.

How it works

In a closed system, evolution is unitary (according to the Schrödinger equation), and a pure state remains pure—entropy does not change. But real systems interact with the environment, leading to decoherence—the density matrix becomes mixed, entropy increases. For a composite system in a pure state, the reduced matrix of a subsystem can have nonzero entropy, which precisely measures the degree of quantum entanglement between the parts. The computation involves diagonalizing ρ and the formula Σ −pᵢ log pᵢ. In quantum computing, monitoring this entropy helps combat noise through error correction codes.

S(\rho) = - \operatorname{Tr} (\rho \ln \rho)

S is von Neumann entropy, ρ is the density matrix of the system (quantum state, generalization of probabilistic description), Tr is the trace operation (sum of diagonal elements of the matrix), ln is the natural logarithm defined for the matrix via its eigenvalues. In the representation ρ = ∑ λ_i |i⟩⟨i|, the formula simplifies to S = −∑ λ_i ln λ_i, where λ_i ≥ 0, ∑ λ_i = 1 are the eigenvalues, interpreted as probabilities.

Open problems & fun facts

Exact computation of entropy for many-body systems requires exponential resources even on quantum simulators.; Direct experimental measurement of von Neumann entropy became possible only for small systems (up to a few qubits).; The fundamental connection with thermodynamic entropy in strongly nonequilibrium quantum processes remains a subject of debate.

💡 The strong subadditivity of von Neumann entropy, proved by Lieb and Ruskai in 1973, looks like a triangle inequality for quantum systems: S(ABC) + S(B) ≤ S(AB) + S(BC) and has profound consequences for the structure of quantum correlations.

No articles yet

von Neumann entropy

History

How it works

History

How it works

History

How it works

Open problems & fun facts

Related articles